ABSTRACT Project 3 A largely unresolved problem of in vivo gene therapy with adeno-associated viruses (AAV) is re-administration of vector, which is often diminished by neutralizing antibodies (NAbs) to the AAV capsid that are generated as a result of the first vector administration. The impact of NAbs on vector delivery is greatest when the vector is infused into the circulation, which is the route of administration being used for liver delivery in the treatment of metabolic diseases that are the subject of this P01. Furthermore, approximately 20% of human populations have NAbs against AAV8 due to natural infections which could interfere with the initial treatment. We propose complementary and potentially synergistic approaches for circumventing the inhibitory effects of NAbs. First, we will create a capsid that is capable of high level transduction of human liver but is serologically distinct from the first generation vector based on AAV8; recent data from the Wilson Lab suggests that this can be accomplished with engineered versions of the previously isolated capsid AAV3B. Populations of AAV3B capsid mutagenized in hypervariable regions will be selected in a mouse which harbors a human liver xenograft and has been passively transferred with human AAV NAbs. Candidate capsids that emerge from these selections will be further characterized in nonhuman primates (NHPs) to identify a second generation vector called AAV3Be which will be evaluated for efficacy in a mouse with a liver xenograft derived from human hoFH hepatocytes. We will also study in NHPs the potential of a short course of immune modulation at the time of AAV8 vector delivery to block the formation of NAbs and to enable re-adminstration with an AAV8 or AAV3Be vector.